good practice benchmarking of the rail infrastructure managers · network rail nr gb benchmarking...

Good practice benchmarking of the rail

infrastructure managers

PRIME 2017 data Benchmarking Report

– Public Version

Report developed under cooperation between PRIME KPI & Benchmarking

Subgroup and European Commission Directorate for Mobility and Transport

Hamburg, 03 May 2019

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Foreword by PRIME Co-Chairs

2

The goal of PRIME members is to provide safe, reliable and

efficient railway infrastructure for transporting people and goods.

The KPI subgroup was set up with the goal to monitor and

benchmark performance and by doing so to strive for better results.

We are pleased that we can share with you the second

benchmarking report prepared by the PRIME KPI subgroup,

covering the years 2012-2017.

For the infrastructure managers, benchmarking helps to understand

where each organisation stands and where there is potential for

improvement. For the European Commission, there is an invaluable

opportunity to receive feedback and to monitor the progress with

respect to EU policy priorities. The KPI subgroup has also set up a

database and IT tool which can be used for analysing the trends

and support management decisions on a daily basis.

The PRIME benchmarking framework is:

• comprehensive – including a selection of indicators covering a

broad range of topics and

• has been developed by the industry itself and focussing on what

is useful from the infrastructure managers' business perspective.

We believe that these two elements have been key features to

ensure its wide support. We promised last year that each next

report would be an improvement. And we are proud to confirm that

compared to the first report, this edition includes a number of new

indicators, more complete dataset, three new participants (in total

15) and is enriched by new analysis. Five infrastructure managers

are in the transitional phase to join. We would like to thank the

PRIME KPI subgroup chair Rui Coutinho from IP Portugal - as well

as the members of this group from 20 organisations and EC for this

outstanding achievement.

We believe that PRIME data and definitions can serve the needs of

a large range of industry experts and policy makers. By measuring

and sharing the results, we aim to demonstrate to wider public that

the rail sector is improving its devoted to improve its service

provision.

Finally, we invite remaining PRIME members to join the

benchmarking framework so that our database and report will

gradually become the most renowned source of complete and

reliable data!

PRIME co-chairs

Elisabeth Werner Alain Quinet

European Commission, SNCF Réseau

DG MOVE

Director of Land Transport Deputy Director General

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction

• Benchmarking results

• Appendix

3

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

This report provides an overview of KPI data and results –

It serves as a starting point for further benchmarking

Purpose of this report (1/4)

What is PRIME?

PRIME was created in 2013 as a cooperation platform between the European Commission and the European Rail

Infrastructure Managers, with the view to facilitate the provision of efficient and effective rail services. PRIME has in

total 39 member organisations and 15 of them have participated in the preparation of in this report.

OBJECTIVE OF PRIME PERFORMANCE BENCHMARKING

The 4th Railway Package (Article 7f of the Directive 2012/34/EU, as amended by Directive 2016/2370) has formalised

and specified the missions of PRIME. In particular, it states that “[…] the network meets at regular intervals to […]

monitor and benchmark performance. For this purpose, the network shall identify common principles and practices for

the monitoring and benchmarking of performance in a consistent manner”.

Infrastructure managers are natural monopolies and performance benchmarking is a relevant exercise to assess,

manage and improve their performance. Many indicators are already available within the sector but they are not

harmonised and are incomplete. Now, for the first time, all Infrastructure Managers are mobilised to provide a

coherent framework of performance indicators.

4

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/




OBJECTIVE OF PRIME PERFORMANCE BENCHMARKING (continued)

Performance Benchmarking covers several dimensions of rail infrastructure management: punctuality, costs,

resilience, sustainable development, safety, etc. Our objective is to provide a comprehensive view of the

performance of the networks with the opportunity for Infrastructure Managers to identify areas for improvement and

the sources of inspiration among their peers.

A second internal benchmarking report has been produced based on 2012-2017 data accompanied by

assessment of data completeness and robustness, of 49 selected indicators and first assessment of KPI correlations,

qualitative relationships between KPIs and potential performance drivers in the different performance dimensions.

The purpose of this report was to illustrate the current performance of IMs and identify areas for further analysis.

Thus, this is only the beginning of a longer term process.

Compared to PRIME 2016 data Benchmarking report published last year, we have already achieved a significant

improvement of the dataset, especially in terms of completeness. Furthermore, we have started to drill down into the

subject of punctuality and developed a separate analysis. Our intention is to give information and fruit for thought to

stakeholders, researchers, economists and politicians. Above all, the general objective for the project is to deliver

insight and inspiration for better decisions on developing a sustainable and competitive infrastructure

management which provides high quality services.

Thanks to the strong commitment of a large number of Infrastructure Managers, we are confident to be able to

progressively improve the participation and the publication with the view to foster accountability, transparency and,

ultimately, performance.

5

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/




OPERATIONAL ACHIEVEMENTS

PRIME KPI and its Benchmarking Subgroup has been working actively for the last five years. Through more than 30

meetings, 15 active member organizations and three pilot projects we have achieved the following results:

• An internal IT tool developed by the EC IT team in cooperation with civity Management Consultants is now

established. Several improvements have been made to increase its usability and functionality.

• The KPI definitions are documented in a next version of a PRIME KPI Catalogue that is available on

https://webgate.ec.europa.eu/multisite/primeinfrastructure/content/subgroups_en

PRIME 2017 BENCHMARKING REPORT: THE STARTING POINT FOR FURTHER BENCHMARKING

• The present PRIME 2017 Benchmarking report shows the results of a selection of indicators which are based

on the initial assessment of the internal report were considered mature enough for publishing. This report with

purely factual information is the second edition, facilitating further data sharing and analysis. As indicated in the

document, for some indicators, the data of individual infrastructure managers partially still deviates from agreed

definitions, but the members continue their efforts to improve the comparability of data.

• This is PRIME’s second Benchmarking report and it shows significant progress compared to the first version of

2016. However, the participating members remain committed that each next report will become an improvement

over the previous one.

6

https://webgate.ec.europa.eu/multisite/primeinfrastructure/content/subgroups_en

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/




PRIME KPI NEXT STEPS

• Enhancing participation: the number of members involved in the benchmarking report, currently 15 will

progressively increase

• Improving the dataset: The KPI framework will continue to be developed over the coming years, with the KPIs

refined, completed, and the quality of the input data and hence output metrics improved

• In-depth studies: based on the results achieved, PRIME will work on in-depth analyses which include

interpretation of benchmarking results with detailed analyses of contextual factors and identification of root causes

for performance differences on selected topics; the topic chosen for 2018 is punctuality

• Preparing and sharing reports: PRIME aims to publish annual benchmarking reports. In addition it will prepare

'special reports' presenting the outcome of the in-depth analyses

7

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

A number of factors need to be in place to make this

benchmarking exercise successful

Context – Key success factors of PRIME KPI

There are a number of factors to be considered for a successful and meaningful benchmarking exercise:

Meaningful and supportive KPIs strongly aligned with the peer group’s strategic objectives and providing a

good starting point for the identification of good practices

Clear and well defined indicators are essential for reliable and comparable results

Reliable and high data quality through a thorough challenging of the collection and completeness of data

including plausibility checks and gap-filling

Comparability of results can be increased by applying adjustments to normalise data based on structural

differences between IMs, as well as identifying limitations and caveats very clearly to avoid misinterpretation

and misleading conclusions

Target group-oriented tools and reporting should be developed which are flexible, easy-to-use and

correspond to the needs of benchmarking experts, team members, and senior managers, etc., using carefully

defined requirements.

A strong senior management commitment is essential to support and resource the exercise, and provide

confidence to interpret, understand and implement results

8

!

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

15 participants contributed to this report - 7 new members

have joined PRIME’s KPI benchmarking subgroup

Context – PRIME KPI active members

9

Participants in PRIME KPI Report New subgroup members in transition phase PRIME members

Observers:

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

IM nameIM

abbreviationCountry Participation

Main

track-kmFTE

Adif Adif ES Benchmarking Report 21.029 12.079

Bane NOR Bane NOR NO Benchmarking Report 4.125 4.420

DB Netz AG DB DE Benchmarking Report 55.311 43.974¹⁾

Finnish Transport Infrastructure

AgencyFTIA FI Benchmarking Report 6.708 639

Infrabel Infrabel BE Benchmarking Report 6.515 11.361²⁾

Infraestruturas de Portugal S.A. IP PT Benchmarking Report 3.244 3.697

Latvijas dzelzceļš LDZ LV Benchmarking Report 2.217 6.494

Lietuvos geležinkeliai LG LT Benchmarking Report 1.911 2.987

Network Rail NR GB Benchmarking Report 31.221 38.594

PKP PLK PKP PLK PL Benchmarking Report 27.120 39.349

ProRail ProRail NL Benchmarking Report 5.409 4.280

RFI RFI IT Benchmarking Report 27.044 25.963

SBB SBB CH Benchmarking Report 6.183 9.450²⁾

SNCF Réseau SNCF R. FR Benchmarking Report 48.992 53.624

Trafikverket TRV SE Benchmarking Report 11.775 7.135

258.805 199.261

The 15 participants manage more than 250 thousand main

track-kilometres and employ about 200 thousand FTE

Context – PRIME KPI active members

10

1) 2016 data

2) 2015 data

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Currently seven organisations are in transition to becoming

a member delivering data

Context – PRIME KPI new members

11

IM nameIM


Banedanmark BDK DK New member in transition

Győr-Sopron-Ebenfurti Vasút GySEV HU/AT New member in transition

HŽ Infrastruktura d.o.o. HZ HR New member in transition

Iarnród Éireann – Irish Rail IE IE New member in transition

LISEA LISEA FR New member in transition

MÁV Magyar Államvasutak Zrt. MÁV HU New member in transition

Správa železniční dopravní cesty SZDC CZ New member in transition

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Eight EU main infrastructure managers do not yet

participate

Context – PRIME members, not active in KPI subgroup

12

IM nameIM


CFR CFR RO PRIME

Eesti Raudtee, AS EE PRIME

National Railway Infrastructure

CompanyNRIC BG PRIME

ÖBB Infrastruktur AG ÖBB AT PRIME

OSE.SA - the Hellenic Railways

OrganisationEL PRIME

Slovenske železnice SI PRIME

Société Nationale des Chemins

de Fer LuxembourgeoisCFL LU PRIME

Zeleznice Slovenskej republiky ZSR SK PRIME

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


• Appendix

13

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

This report provides all high level and benchmarking KPIs

of the framework's business dimensions

Introduction (1/2)

• First, the hierarchy of all KPIs is illustrated followed by an example slide of results explaining

contents and meaning of the graphical illustration

• Each business dimension is introduced by its objectives as described in the PRIME catalogue

• Each category is introduced by a description of the current definitions of its associated KPIs

• This is followed by a comparison of these KPIs per IM illustrated in bar-charts showing for each

IM the most recent available data among the years 2012 – 2017. Where KPI values for 2017 are

currently not available, KPI values are based on data from the most recent available year. For

example, if the latest data provided by an IM is from 2016 then this 2016 data is presented in the

bar chart

• Bar-charts also indicate the weighted average across all IMs (weighted by the KPI’s denominator)

based on most recent available data as well as the individual IM’s averages (over all available

years 2012 – 2017)

• The result of each comparison is described emphasizing the average, the range and individual

trends where meaningful

• As requested by the KPI subgroup, benchmarking results are not interpreted and possible reasons

for performance differences are not investigated in detail at this stage

• Instead first questions for further analysis are raised

• For better readability, bar charts are not labelled with individual values

14

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

This report provides all high level and benchmarking KPIs

of the framework's business dimensions

Introduction (2/2)

• A comparison of individual time series is not illustrated in order to keep up readability; however,

time series are available in the IT-tool

• Where relevant and helpful to support further analysis, first correlations between KPIs are

illustrated and commented

• In order to identify exogenous and endogenous drivers of performance differences, each business

dimension (except for context) is concluded by

– A graphical illustration of examples for underlying drivers developed by civity compared to the

KPIs currently collected in PRIME

– A summary of possible guiding questions for further analysis

• This first root-cause analysis can be used to explain performance differences as well as to identify

possible in-depth topics for the KPI subgroup

15

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Performance indicator hierarchy

The PRIME performance indicators have been tiered into

four levels, with the main KPIs presented in this report

Tier Indicators Reporting

High Level

Industry KPIs

Bench-

marking

KPIs

Additional

PIs

Supporting

Indicators & data

Selection of 12 top KPIs

Additional 32 KPIs covering

all categories for core

benchmarking

All remaining PIs and

KPIs under review

Other indicators & data for

detail and explanation

Bench-

marking

report

Report-

ing per

indicator

in IT tool

16

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

High level and benchmarking KPIs

17

PRIME

Context

Electrification

Modal share

passenger transport

Modal share freight

transport

Safety

Accidents

Precursors

Fatalities

Security

Delays

Train cancellations

Environment

Diesel trains1)

Electric trains1)

CO2 emissions

Capacity

Possessions planned

Possessions utilised

Condition

Asset failures

Signalling

Telecom

Power supply

Track

Structures

Other

Permanent speed

restrictions

Temporary speed

restrictions

Costs

OPEX

Maintenance

Traffic management

CAPEX

Renewals

Revenues

Incentive regimes

Utilisation

Train-km

Passenger trains

Freight trains

Asset Capability & ERTMS

Deployment today

Deployment 2030

Intermodality

Intermodal stations

Passengers at

accessible stations

Context Safety & Environment Delivery Financial Growth

Punctuality

Passenger trains

Freight trains

Delays caused by IM

Train cancellation

caused by IM

Reliability

Delays

Signalling

Telecom

Power supply

Track

Structures

Other

Performance

The KPIs presented in this report include 12 high level

industry and 32 benchmarking KPIs across six dimensions

High Level Industry KPI Benchmarking KPI

Non access charges

Track access charges

Proportion

KPI under review

116

117

43

44

51

52

53

54

55

56

57

58

59

60

62

66

68

64

87

81

91

80

35

36

37

38

39

40

41

28

29

31

34

18

19

20

7

12

8

15

16

92

93

94

98

101

1

2

3

1) For the purpose of this report “Share of train types” (combination of KPI 18 & 19) is considered as a high level KPI

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Source: civity calculations using data as provided by the infrastructure managers until 29 January 2019Data accuracy: No entry = Normal E = Estimate D = Deviating from definition P = Preliminary

Latest available year Average of available years 2012-2017 Total weighted average of each IMs latest available year

(accu-

racy)IM • Delays vary between 2 and 17 minutes per

thousand track-kilometres. ProRail shows a

significantly lower level of delay minutes than

others and this is consistent with its good

overall passenger punctuality (KPI 28)

• Several IMs have considerably reduced

delays in their responsibility in 2017

• It would be interesting to

– break down delay minutes by cause to

identify main causes, positive trends and

thus opportunities for reducing delays

– understand the reasons and initiatives

behind positive trends

Example slide of results

18

Delay minutes per train-km caused by the IM

Minutes per thousand train-km (2017)

KPI 31

Delay causes include: Operational

planning, Infrastructure installations, Civil

engineering causes, Causes of other IM6,47

0 5 10 15 20

Adif

Bane NOR

DB

FTIA (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV

The traffic light indicates the overall

robustness level (see appendix) of the KPI

Red dot indicates the average

value of all available years

If neither grey bar nor

red dot visible: no data

delivered for any year

Grey bar indicates

the value for the

latest available year

Name, unit and

year of most

recent data of KPI

Dotted line indicates the

weighted average of all

latest available years

(weighted by the KPI’s

denominator)

Infrastructure Manager,

footnote indicating year

if not current year

Accuracy level of data,

empty if Normal

Number of KPI, colour

coding indicates KPI level

(compare High level and

benchmarking KPIs

overview on previous page)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context

– Safety and environment

– Performance

– Delivery

– Financial

– Growth

• Appendix

19

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Understanding the size and relative significance of the

railway in each country and the market for railway services

• Provision of valuable background information and relevant

context when reviewing and assessing other KPIs and

additional performance indicators

This category provides an overview of the characteristics

and configuration of each IM

Context – objectives

20

Source: PRIME Catalogue Version 2.1, 31 May 2018

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

This category provides an overview of the characteristics and configuration of each IM. This enables an understanding of the size and relative significance

of the railway in each country and the market for railway services, which provides valuable background information and relevant context when reviewing

and assessing other KPIs and additional performance indicators.

KPI DefinitionKPI Name

Context – Overview

21

PRIME

Context Safety & Environment Performance Delivery Financial Growth

Degree of electrification of total network - all lines

Proportion of national rail passenger-kilometre compared to total passenger-kilometre of passenger cars, buses / coaches,

and railways (Source: European Commission, Statistical Pocket book)

Proportion of national rail tonne-kilometre compared to total tonne-kilometre of road, inland waterways and rail freight

(Source: European Commission, Statistical Pocket book)

Degree of electrification of

total network – all lines

National modal share of rail in

passenger transport

National modal share of rail in

freight transport

Context

Electrification

Modal share

passenger transport

Modal share freight

transport

1

2

3

High Level Industry KPI Benchmarking KPI KPI under review

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

67% of the peer group’s track-kilometres are electrified

22

Degree of electrification of total network

% of track-km (2017)

KPI 1

• In Europe railway networks are mostly

electrified

• However, the degree of electrification varies

strongly from 9% to 100%

• Infrabel, Trafikverket and SBB have the

highest degree of electrification

• Overall the degree of electrification has been

quite stable in the period considered

NR: currently only report main electrified

track-km and is exploring further

categorisation into electrified total and

main track-km66,8

0 50 100

Adif

Bane NOR

DB

FTIA

Infrabel (E)

IP (E)

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Based on passenger-kilometres, the peer group’s

average modal share of rail in passenger transport is 7%

23

National modal share of rail in passenger transport

% of passenger-km (2016)

• The range of national modal shares varies

widely between 1% and 17%

• The highest modal share of passenger rail

transport can be found in Switzerland (17%)

• With a few slight exceptions, modal shares

appear to be relatively constant over time

KPI 2

• Data provided by European

Commission

• Source: Eurostat based on data

reported by national statistical offices

• 2017 data will only become available

during the course of 20197,45

0 5 10 15 20

ES

NO

DE

FI

BE

PT

LV

LT

UK

PL

NL

IT

CH

FR

SE

Country

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Based on tonne-kilometres, the peer group’s average

modal share of rail in freight transport is 24%

24

National modal share of rail in freight transport

% of tonne-km (2016)

• The range of national modal shares varies

widely between 5% and 77%

• The highest modal share of freight rail

transport can be found in Latvia (77%)

• All modal shares appear to be relatively

constant over time except for a slight

decrease in a few countries

KPI 3

24,2

0 50 100

ES

NO

DE

FI

BE

PT

LV

LT

UK

PL

NL

IT

CH

FR

SE

Country

• Data provided by European

Commission

• Source: Eurostat based on data

reported by national statistical offices

• 2017 data will only become available

during the course of 2019

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context


– Performance

– Delivery

– Financial

– Growth

• Appendix

25

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Understand and improve the ability of an IM to manage and

operate its network and users of its network in such a way

as to maximise safety and security (ALARP) for its

customers, staff, its partners – operators, contractors and

suppliers – and the general public; and

• Demonstrate the ability of an IM to manage its network in

such a way as to minimise short term and long term

environmental impacts by itself and its staff, its operators,

suppliers and customers.

Aim is to demonstrate the level of safety and security as

well as the environmental impact provided by the railway

Safety, Security & Environment – objectives

26


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Safety is the primary focus of the management of a railway IM and a prerequisite in any framework of management indicators. It is the most important and

essential element in the performance of an IM, and affects customers, stakeholders, the reputation of the IM, the railway and society at large.

Safety & Environment – Safety – Overview

27

PRIME


Relative number of significant accidents including sidings, excluding accidents in workshops, warehouses and depots

based on the following types of accidents (primary accidents): Collision of train with rail vehicle, Collision of train with

obstacle within the clearance gauge, Derailment of train, Level crossing accident, including accident involving

pedestrians at level crossing, Accident to persons involving rolling stock in motion, with the exception of suicides and

attempted suicides, Fire on rolling stock, Other accident

The boundary is the point at which the railway vehicle leaving the workshop / warehouse / depot / sidings cannot pass

without having an authorization to access the mainline or other similar line. This point is usually identified by a signal. For

further guidance, please see ERA Implementation Guidance on CSIs.

Relative number of the following types of precursors: broken rail track buckle and track misalignment wrong-side

signalling failure

Relative number of persons seriously injured (i.e. hospitalised for more than 24 hours, excluding any attempted suicide)

and killed (i.e. killed immediately or dying within 30 days, excluding any suicide) by accidents based upon following

categories: Passenger, Employee or contractor, Level crossing user, Trespasser, Other person at a platform,

Other person not at a platform

Significant accidents

IM related precursors to

accidents

Persons seriously injured and

killed


Security

Delays

Train cancellations

Environment

Diesel trains

Electric trains

CO2 emissions

Safety

Accidents

Precursors

Fatalities

18

19

20

7

12

8

15

16


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average the peer group’s infrastructure networks

show 0,3 significant accidents per million train-kilometres

28

Significant accidents

Number per million train-km (2017)

• Both in recent years and on average over

available years ProRail, NR and SBB operate

its railways at the lowest accident levels

• In contrast a few IMs significantly exceed the

weighted average

• Further analysis should explore the root

causes of accidents and possible mitigation

measures

• Good practice of IMs to improve their overall

safety performance (reduce the number of

accidents and accident precursors) could be

evaluated further, e.g. programmes to in-

crease safety at level crossings, track worker

safety or the safety level of signalling systems

• Given that metrics include accidents

exclusively due to train operation, a further

breakdown and in-depth analysis may be

needed

KPI 7

0,32

0 1 2

Adif

Bane NOR

DB (D)

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

The number of persons seriously injured and killed

during the reporting period varies widely

29

Persons seriously injured and killed


KPI 8

• The weighted average of safety related

injuries and fatalities in the peer group's

railway network is 0,3 per million train-

kilometres

• They are lowest at ProRail in 2017 at 0,11;

NR maintains the lowest average over time

• The casualty rate on some networks are well

above the weighted average

• As safety is the most crucial aspect in

delivering railway services it is worth to

understand how best practice can be

achieved

• Hence further analysis could consider :

– Which were types of accidents and their

underlying causes?

– What technical measures, regulation or

other measures are taken to further

increase safety levels?

0,32

0 1 2

Adif

Bane NOR

DB (D)

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Precursors like broken rails and wrong side signalling

failures occur 1,6 times per million train-kilometres

30

IM related precursors to accidents


KPI 12

• Precursors are a good indicator to

understand and mitigate root causes for

significant accidents (for example the number

of train buckles leading to a risk of train

derailments)

• The number of precursors of the peer group

varies widely, some showing levels well below

the peer group’s weighted average while

others have significantly higher values

• For a further analysis a breakdown by

precursor and the underlying reasons would

be valuable; it is also of interest to

understand the impact/severity of different

precursors

Initial definition was to collect all precursor

data. The definition was then narrowed

down to a few precursors.1,60

0 2 4 6 8

Adif

Bane NOR

DB (E)

FTIA

Infrabel

IP

LDZ

LG

NR¹⁾PKP PLK

ProRail

RFI

SBB

SNCF R.¹⁾TRV

1) Data of 2016

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The management of railway security includes activities for the protection of the railway, its users and its staff through monitoring, prevention and

preparation of responses to security incidents carried out with malicious intent, which have the potential to harm customers and staff, damage railway

assets, or generally to impede and disrupt railway operations.

Safety & Environment – Security – Overview

31

PRIME


Number of delay minutes due to security incidents (intentional acts as terrorism, sabotage, cyber-attacks, vandalism, thefts,

espionage, unauthorized persons and other acts of aggression or hooliganism) per train-kilometre

Percentage of trains cancelled caused by security incidents (intentional acts as terrorism, sabotage, cyber-attacks,

vandalism, thefts, espionage, unauthorized persons and other acts of aggression or hooliganism) per total trains scheduled

to be operated

Delays caused by security

incidents

National Train cancellations

caused by security incidents


Security

Delays

Train cancellations

Environment

Diesel trains

Electric trains

CO2 emissions

Safety

Accidents

Precursors

Fatalities

18

19

20

7

12

8

15

16


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average security incidents cause 1,3 delays minutes

per thousand train-kilometres

32

Delays caused by security incidents


KPI 15

• The KPI still has a small dataset with an

increase in data provision in 2017

• Data shows a range of 0,2 to 2,4 delay

minutes per thousand train-kilometres caused

by security incidents

• There seem to be some dynamics over the

years as average values are quite different

compared to values of 2017

1,34

0 1 2 3

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Monitoring the environmental impact of the IM focuses on two aspects: the influence of the IM in affecting and improving the environmental impact of the

whole integrated railway (e.g. through electrification) and the direct environmental impact of the IM’s own activities.

Safety & Environment – Environment – Overview

33

Diesel train-kilometres compared to train-kilometres both for passenger and freight trains

Electric train-kilometres compared to train-kilometres both for passenger and freight trains

CO2 emission produced from maintenance rolling stock compared to main track-kilometre

Share of diesel trains

Share of electric trains

Performance against carbon

reduction target


Security

Delays

Train cancellations

Environment

Diesel trains

Electric trains

CO2 emissions

Safety

Accidents

Precursors

Fatalities

PRIME


18

19

20

7

12

8

15

16


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The majority of train-kilometres in the peer group results

from electricity-powered trains

Share of train types1)

% of total train-km (2017)

34

Source: civity calculations using data as provided by the infrastructure managers until 29 January 2019

0 10 20 30 40 50 60 70 80 90 100

Adif

LDZ

DB

Bane NOR

FTIA

Infrabel

LG

IP

NR

PKP PLK

TRV

ProRail

RFI

SBB

SNCF R.

77

Share of electricity-powered trains Work trainsShare of diesel-powered trains Unknown

KPI 18+19

• Overall the share of electrically produced

train-kilometres in the peer group is quite

high, reaching 77% of the total

• This reflects the degree of electrification of

the network which for most organisations

reaches 70% or more (KPI 1)

• The weighted average of the peer

group is drawn down particularly by

NR’s high reliance on Diesel engines

• Unknown share for Adif, FTIA and NR

are likely to refer to work trains

Total weighted average of electricity-powered trains

1) For the purpose of this report “Share of train types” (combination of KPI 18 & 19) is considered as a high level KPI

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Share of electricity-powered trains / Electrification

35

45

50

55

60

65

70

75

80

85

90

95

100

45 50 55 60 65 70 75 80 85 90 95 100

IP

Adif

ProRailInfrabel

1: Degree of electrification of total network - main lines

(% of track-km)

19: Share of electricity-powered trains

(% of train-km)

FTIA

Bane NOR

PKP PLK

RFI

SBB

SNCF R.

TRV

As expected there is a strong correlation between the

degree of network electrification and share of electric trains

• In general there is a correlation between the

degree of electrification and the share of

electric train-kilometre produced in the

network

• However it is noticeable that similar degrees

of electrification do not lead to similar shares

of electrically produced train services

• For instance, SBB reaches a significantly

higher share of 99% while other IMs achieve

less than 90% although the degree of

electrification is higher

• It should be further explored why this is the

case and if there are opportunities to more

extensively use the electrified infrastructure

by reducing the share of diesel trains

• Aspects of further analysis could be:

electrification strategy, utilisation of lines,

coordination with fleet investments

• LG (10%/9%) and LDZ (21%/16%) were

excluded for readability reasons

KPI 19 / 1

2017

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average IMs’ maintenance rolling stock emits 0,6

tonnes of CO2 per main track-kilometre

36

CO2 emission produced from maintenance rolling stock

tCO2 per main track-km (2017)

KPI 20

• The environmental impact of an IM’s

maintenance rolling stock is measured by its

CO2 emissions

• On average 0,6 tonnes are annually emitted

per main track-kilometre

• However, there are quite large differences

between IMs’ reporting data which should be

further investigated

• Relevant questions are around

– The intensity of use of this fleet

– The amount of fleet operated by the IM

and considered here (in contrast to fleet

operated by contractors)

– The structure of the fleet in use

0,59

0 0,5 1 1,5 2

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Safety & Environment – drivers1)

37

Further analysis on safety and environment could be based

on the set of the drivers illustrated below

PRIME KPIs2)

Safety

Accidents

Precursors

Fatalities

Security

Delays

Train cancellations

Environment

Diesel trains

Electric trains

CO2 emissions

Safety & Environment

Train safety

Safety &

Environment

Track worker

safety

Network

operations

Environmental

impact+ ++

Security in

stations and

buildings

+

Condition of rolling stock

Monitoring systems

(Automated) Warning systems

Safety education

Measure against trespassing (e.g. fencing)

Level crossing density

Signalling failures

Infrastructure conditions

Degree of elec-trification (KPI 1)

Electrified trains (KPI 19)

Carbon footprint (CO2 emissions) (KPI 20)

Energy consumption

Access control systems

Monitoring systems

1) Drivers which are currently collected in PRIME are coloured light blue

2) As currently collected and evaluated in PRIME

Responsibility of

Train Operating

Companies

Security staff

ETCS

18

19

20

7

12

8

15

16

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Further analysis should account for external performance

drivers and identify opportunities for improvement

Safety & Environment – Further analysis

• In order to improve safety performance (reduce the number of accidents and accident

precursors) it would be valuable to investigate the root causes and the programmes that IMs

initiated to mitigate them

• As precursors to accidents are an important indicator when working on the prevention of critical

accidents their impact and severity should be explored more in detail; it would also be valuable to

understand how IMs reduce both the number of precursors and the resulting number of

significant accidents

• In the area of security the definition of the KPI “train cancellations caused by safety incidents”

needs to be improved and finalised as this KPIs is still critical

• Electrification and the use of electrified trains are important environmental indicators; it should be

further explored why levels of utilisation of electrified track are different and how electrified

track can be exploited better

• The use of maintenance rolling stock needs a deeper analysis in order to understand where the

differences result from, e.g. considering the volume of maintenance fleet and the intensity of use

• It would be very valuable if well performing IMs and those who improve over time reported on

practices and initiatives which contribute to their safety and environmental performance

38

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context


– Performance

– Delivery

– Financial

– Growth

• Appendix

39

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Understand the performance of the IM network in relation

to other IMs;

• Improve the ability of the IM to enable trains to run on time;

and,

• Identify opportunities to improve the management of assets

to minimise the number of failures, and the impact of those

failures on the operating railway.

Aim is to describe the network performance and the

resulting impact on operators and customers

Performance – objectives

40


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Train punctuality is the primary measure of overall railway performance and a key measure of quality of service, driven not only by the IM but also opera-

tors and customers. The requirements for punctuality differs between IMs, high-speed routes, core network, customer groups, passenger/freight etc. It is

essential to understand both the overall performance of the system through punctuality, as well as the IM’s impact on and responsibility for punctuality.

Performance – Punctuality – Overview

41

PRIME


For national and international passenger trains (excluding work trains) the percentage of all trains which arrive at each

measuring point with a delay of less than or equal to 5:29 minutes compared to all trains actually operated (i.e. were not

cancelled) out of those that were scheduled in the original working timetable, including those timetabled at short notice

Passenger trains punctuality

For national and international freight trains (excluding work trains) the percentage of all trains which arrive at each

measuring point with a delay of less than or equal to 15:29 minutes compared to all trains actually operated (i.e. were not

cancelled) out of those that were scheduled in the original working timetable, including those timetabled at short notice

Freight trains punctuality

Average delay minutes per train-km caused by incidents that are regarded as IMs responsibility according to UIC leaflet

450-2, Appendix A - Table 1 (columns 1 through 3) and Appendix B.1 through B.3. Train-kms considered are total train-km

operated (revenue service + shunting operations to and from depots + IM’s work traffic). Delay minutes will be measured at

all available measuring points. Of these measured delay minutes the maximum number is counted if it exceeds a threshold

of 5:29 minutes for passenger services and 15:29 minutes for freight services. This is in accordance with UIC leaflet 450-2

chapter 4.1 - Rounding rules, number 2. No delay minutes are counted if these thresholds are not exceeded at any

measuring point

Delay minutes per train-km

caused by the IM

Percentage of fully or partially cancelled national and international passenger trains that are included in the last working

timetable issued the day before the service (or the timetable that is valid when the train service takes place) which were

caused by incidents that are regarded as IMs responsibility according to UIC leaflet 450-2, Appendix A - Table 1 (columns 1

through 3) and Appendix B.1 through B.3. All four types of cancelled trains are to be included: full cancellation (cancelled at

origin), part cancellation en route, part cancellation changed origin, part cancellation diverted

Percentage of train

cancellations caused by the

IM


Punctuality

Passenger trains

Freight trains

Delays caused by IM

Train cancellation

caused by IM

Reliability

Delays

Signalling

Telecom

Power supply

Track

Structures

Other

35

36

37

38

39

40

41

28

29

31

34


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

90,7

80 85 90 95 100

Adif (D)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ (D)

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV (E)

• Further work is undertaken by IMs to collect

punctuality data according to the PRIME

definition, in order to make this measure more

comparable across the peer group

• Among IMs with normal data SBB and ProRail

show highest levels of punctuality. FTIA and IP

have more delays compared to last years’

average

• It would be interesting to analyse:

– reasons behind the good and improving

performances of individual IMs

– external drivers of performance differences

such as utilisation or network complexity

On average 91% of passenger trains are on time

42

Passenger trains punctuality

% of trains (2017)

KPI 28

Some IMs use differing observation points

and rounding rules for measuring

punctuality

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average 68% of freight trains are on time

43

Freight trains punctuality

% of trains (2017)

• Further work is required by IMs to collect

punctuality data according to the PRIME

definition, in order to make this measure more

comparable across the peer group

• Among the IMs with normal data, freight

punctuality is highest for Bane NOR, FTIA and

SBB. Freight punctuality varies by a factor of 2

and is considerably lower than for passenger

traffic, despite its higher delay threshold

• It would be interesting to understand

– why there is such a wide variation in freight

train punctuality

– the reasons behind the good performances

KPI 29

Some IMs use differing observation points

and rounding rules for measuring

punctuality68,2

0 50 100

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ (D)

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV (E)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM • Delays vary between 2 and 9 minutes per

thousand train-kilometre. ProRail shows a

significantly lower level of delay minutes than

others and this is consistent with its good

overall passenger punctuality (KPI 28)

• Several IMs have considerably reduced

delays in their responsibility in 2017

• It would be interesting to

– break down delay minutes by cause to

identify main causes, positive trends and

thus opportunities for reducing delays

– understand the reasons and initiatives

behind positive trends

The average of delays caused by IMs is 6 minutes per

thousand train-kilometre

44

Delay minutes per train-km caused by the IM


KPI 31

Delay causes include: Operational

planning, Infrastructure installations, Civil

engineering causes, Causes of other IM6,47

0 5 10 15 20

Adif

Bane NOR

DB

FTIA (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM • The percentage of train cancellations caused

by IMs varies widely, some showing levels

well below the weighted average while others

have significantly higher values

• IP stands out as causing no or very few train

cancellations. The main reason is that IP had

very few infrastructure related works leading

to cancellations

• It would be interesting

– for IP to provide explanations for their

outstanding performance

– to understand the relationship between

delays and train cancellations caused by

IMs

– to understand the breakdown of train

cancellations by cause (type of incidents

within IM’s responsibility)

On average IMs cause 29 percent of all passenger train

cancellations

45

Passenger train cancellations caused by the IM

% of scheduled and cancelled passenger trains (2017)

KPI 34

28,7

0 20 40 60 80

Adif

Bane NOR

DB

FTIA (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV (D)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Reliability of the infrastructure demonstrates the impact of failures. As well as managing its assets in such a way as to minimise the effect of failures on

the railway, these indicators also measure the effectiveness and timeliness of the IM in responding to these failures, and returning the network to normal

function.

Performance – Reliability – Overview

46

PRIME


Average delay minutes caused by asset failures on main track according to UIC CODE 450-2, numbers 20-25 and 28-29.

Delay causes should include both primary causes and secondary causes.Average delay minutes per

assets failures

… numbers 20 & 21 including failures related to signalling installations and signalling installations at level crossings.

Average delay minutes per …

failures

… number 22 including failures related to Telecommunications (GSM-R, Radio failure and more).

… number 23 including failures in the power supply for electric traction, others and variation and drops of voltage.

… number 24 including failures due to rail breakage, lateral distortion and other track failures.

… number 25 including failures at bridges and tunnels.

… number 28 & 29 including failures according to the managing and planning of staff and other failures.


Punctuality

Passenger trains

Freight trains

Delays caused by IM

Train cancellation

caused by IM

Reliability

Delays

Signalling

Telecom

Power supply

Track

Structures

Other

35

36

37

38

39

40

41

28

29

31

34


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average asset failures cause a delay of 57 minutes

47

Average delay minutes per asset failure

Minutes per failure (2017)

• The average delay minutes per asset failure

varies widely

• Further work is required by IMs to collect data

according to the PRIME definition, in order to

make this analysis meaningful.

KPI 35

57,2

0 100 200 300

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail (D)

RFI (D)

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Across the peer group, power supply, structure and tele-

com failures have particularly large avg. impacts on delays

Average delay minutes per asset failure


48

53 58

88

103

155

175

Power

supply

Signalling StructuresTelecomOther

infrastructure

Track

• Power supply failures have the largest

average impact on delays with 175 minutes

per failure followed by structure failures with

155 minutes per failure. However Structure

failures have to lowest occurrence by a large

margin

• These are followed by average delay minutes

per telecommunication failure (103 minutes)

and track failure (88 minutes)

• The average impacts of other failures (58

minutes) and signalling failures (53 minutes)

are comparatively low, however signalling

failures are the most frequent by far

• The frequency of failures in these asset

groups needs to be considered in order to

determine the overall impact on punctuality

KPI 35

Peer average

7 7 7 7 7 7 IMs who delivered1)

50.611 9.157 9.359 1.871 96 5.711 # of asset failures

66 12 12 2 0 7 % of asset failures

1) both asset failures and delay minutes for each asset class

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average a signalling failure causes a delay of 50

minutes

49

Average delay minutes per signalling failure




make this comparative analysis meaningful

• It may be acknowledged that PKP PLK show

levels significantly below the weighted

average

• It would therefore be valuable for PKP PLK to

provide explanations of the work that they

have implemented to achieve low and

decreasing levels of average delays

• Signalling failures have a relatively low

average impact on delays when compared to

other types of failures (KPI 35) but they

account for 66% of all asset failures recorded

by the peer group (KPI 51)

KPI 36

50,4

0 50 100 150

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average a telecommunication failure causes a delay

of 84 minutes

50

Average delay minutes per telecommunication failure





• Nevertheless it may be acknowledged that

PKP PLK manage to limit the impact of

telecom failures to single digit delays

• While telecom failures account for only 2% of

all asset failures recorded by the peer group

(KPI 51) the average impact of a telecom

failure on delays is high when compared to

the impact of other types of failures (KPI 35)

KPI 37

84

0 100 200 300 400

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average a power supply failure cause a delay of 153

minutes

51

Average delay minutes per power supply failure





• It may be acknowledged that PKP PLK

manage a significantly lower level of delays

per failure

• While power supply failures account for only

7% of all asset failures recorded by the peer

group (KPI 51) a power supply failure causes

one of the largest average impacts on delays

among all types of asset failures, according to

this sample

KPI 38

153

0 200 400 600

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average a track failure causes a delay of 88 minutes

52

Average delay minutes per track failure





• The track system accounts for 12% of all

asset failures recorded by the peer group

making it the second least reliable asset

group behind the signalling system (KPI 51)

KPI 39

87,6

0 100 200 300

Adif

Bane NOR (E)

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average a structure failure causes a delay of 155

minutes

53

Average delay minutes per structure failure





• IP and LG did not record any structure

failures in 2017

KPI 40

155

0 100 200 300 400

Adif

Bane NOR (E)

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

No structure failures occurred

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average any other failure causes a delay of 48

minutes

54

Average delay minutes per other failure





• Nevertheless it may be acknowledged that

PKP PLK manage a significantly lower level

of delays per failure

• It would then be valuable for PKP PLK to

provide explanations of the work that they

have implemented to achieve their outcomes

KPI 41

47,8

0 200 400 600

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Performance – drivers1)

55

IMs are encouraged to use civity's first draft of a root-cause

analysis as basis for discussing performance differences

Network

characteristic

Performance

Network

availability

Traffic

managementExternal factors+ ++ Train operation+

Train frequency (KPI 92)

Complexity of traffic

Planned unavail-ability (maintenan-ce, etc.) (KPI43)

Speed restrictions (KPI 58/59)

Human factor

Time tabling

Decision making

Strike

Natural causes

Administrative formalities

Train preparation

Dwell times

Complexity of the network

Asset reliability (failures) (KPI 51)

MTTR

Rolling stock reliability (failures)

Staff

Punctuality

Passenger trains

Freight trains

Delays caused by IM

Train cancellation

caused by IM

Reliability

Delays

Signalling

Telecom

Power supply

Track

Structures

Other

Performance

PRIME KPIs2)



Mostly manageable factors

Passenger behaviour

Accidents (KPI 7)

Design speed

Asset condition

35

36

37

38

39

40

41

28

29

31

34

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Further analysis should account for external performance

drivers and identify opportunities for improvement

Performance – Further analysis

• In order to understand performances and performance differences between IM networks it

would be good to identify, analyse and account for external drivers of these differences.

• civity’s first draft of a root cause analysis for performance indicates what these external factors are.

For example, it would be helpful to understand the impact of utilisation and network complexity on

punctuality levels

• In order to then identify and assess room and opportunities for improvement, it is

recommended to break down delays (and train cancellations) by cause as well as to analyse the

evolution of this over time. This would allow to identify main delay causes, positive trends and thus

opportunities for reducing delays

• The coverage of delay causes should be improved. Further work is required by IMs to collect

data on average delays per type of failure (KPIs 35-41), construction work (KPIs 43-44) and speed

restrictions (KPIs 58-59) as this data is currently less complete and robust than the majority of

other KPIs

• Having identified potential areas for improvement, it needs to be analysed how performance can

be improved in these areas. A first root cause analysis for performance indicates drivers which are

internally manageable by IMs

• It would be very valuable if well performing IMs and those who improve over time reported on

practices and initiatives which contribute to their performance.

56

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context


– Performance

– Delivery

– Financial

– Growth

• Appendix

57

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Deliver an available, operable and fully functional network,

to the required level of capacity;

• Carry out its asset management functions effectively and in

a timely manner; and

• Maintain and improve asset condition in line with its

strategy.

Aim is to describe the effectiveness of the IM's internal

processes and management of the assets

Delivery – objectives

58


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The Capacity category measures the overall constraints on capacity of the IM’s network. It includes the impact on capacity from the condition of the IM’s

infrastructure and the impact of activities undertaken to maintain or improve overall condition.

Delivery – Capacity – Overview

59

PRIME


Share of main track planned for IMs activities, including maintenance, enhancement and renewals on main tracks. Planned

work in the yearly timetable. This is calculated as the number of main track-km planned for IMs activities weighted by

duration and divided by the total network length

Ratio of executed to planned possessions for IMs activities included in the yearly timetable, including maintenance,

enhancement and renewals on main tracks. This is calculated as the sum of main track-km-days divided by sum of main

track-km-days planned

Possessions planned



Capacity

Possessions planned


Condition

Asset failures

Signalling

Telecom

Power supply

Track

Structures

Other

Permanent speed

restrictions

Temporary speed

restrictions

43

44

51

52

53

54

55

56

57

58

59


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The measurement of asset condition is complex, and not always straight forward for a single IM, never mind as a comparative metric for use in bench-

marking. Therefore the PRIME condition category describes the condition of the asset primarily in terms of how well it functions (i.e. number of failures)

and in terms of the impact of condition of the assets on the expected delivery of the network, in terms of temporary and permanent speed restrictions.

Delivery – Condition – Overview

60

PRIME


Number of asset failures on main track according to UIC CODE 450-2, numbers 20-25 and 28-29 per thousand main track-

km.

Percentage of tracks with permanent speed restriction due to deteriorating asset condition weighted by the time the

restrictions are in place (included in the yearly timetable), related to total main track-km

Percentage of tracks with temporary speed restriction due to deteriorating asset condition weighted by the time the

restrictions are in place (not included in the yearly timetable), related to total main track-km

Assets failures per thousand

main track-km

Tracks with permanent speed

restrictions

Tracks with temporary speed

restrictions

… numbers 20 & 21 … . Including failures related to signalling installations and signalling installations at level crossings.

… failures per thousand main

track-km

… number 22… . Including failures related to Telecommunications (GSM-R, Radio failure and more).

… number 23 … . Including failures in the power supply for electric traction, others and variation and drops of voltage.

… number 24 … . Including failures due to rail breakage, lateral distortion and other track failures.

… number 25 … . Including failures at bridges and tunnels.

… numbers 28 & 29 … . Failures according to the managing and planning of staff and other failures.


Capacity

Possessions planned


Condition

Asset failures

Signalling

Telecom

Power supply

Track

Structures

Other

Permanent speed

restrictions

Temporary speed

restrictions

43

44

51

52

53

54

55

56

57

58

59


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average 900 assets are failing per thousand main

track-kilometre and year

61

Asset failures in relation to network size

Number per thousand main track-km (2017)

• Asset failure frequency in the peer groups’

railway networks varies between 400 and

1.500 failures per thousand main track-

kilometre and year

• Three IMs (BaneNOR, ProRail and SNCF-

Réseau) achieve a failure rate well below the

weighted average

• All failure rates appear to be relatively

constant over time

• Balance between preventive and corrective

maintenance regimes need to be taken into

account

• Extent of use of different failure registration

tools might have an impact on this

comparative analysis

KPI 51

927

0 500 1.000 1.500 2.000

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The majority of asset failures occurs in the signalling

system

Asset failures in relation to network size


62


0 200 400 600 800 1.000 1.200 1.400 1.600

DB

Bane NOR

Adif

FTIA

IP

Infrabel

LDZ

LG

NR

PKP PLK

ProRail

RFI1)

SBB

SNCF R.

TRV

927

Signalling Telecommunication StructuresPower supply Track Other infrastructure Residual asset failures

1) Disaggregation not available

• Based on current data, signalling accounts for

66% of all asset failures

• The track system with 12% is the second

highest failing asset group

• Power supply (7%) and telecommunication

assets (2%) appear to be more reliable

• Structure failure frequency is negligible (0,1%)

• Of course the impact of failures on train

operations is expected to show a different

distribution among the asset groups

• The distribution of other asset failures is very

heterogeneous. Where it is reported, its

impact should not be neglected

KPI 51

Total weighted average

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for signalling assets is 580 per

thousand main track-kilometre and year

63

Signalling failures in relation to network size


• The failure frequency varies widely between

230 and 1.050 failures per thousand main


• Signalling failure rates appear to be relatively

constant over time

• As signalling accounts for the majority of all

asset failures (66%) it would be useful to

identify the most critical components among

all signalling assets

• Also different signalling technologies should

be taken into account

KPI 52

581

0 200 400 600 800 1.000 1.200

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI (D)

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for telecommunication assets

is 24 per thousand main track-kilometre and year

64

Telecommunication failures in relation to network size


KPI 53

24,4

0 50 100 150

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

• The failure frequency varies widely between 2

and nearly 110 failures per thousand main


• Six out of ten IMs (BaneNOR, Infrabel, IP,

ProRail, SNCF R. and TRV) appear to have

quite reliable systems

• Failure rates appear to be relatively constant

over time except for TRV, that shows a

decrease in 2017 compared to the average of

2012-2017

• Even if telecommunication plays a minor role

(2%) in all asset failures, it would be worth to

understand

– The main reasons for failing

telecommunication assets

– What different telecommunication

technologies are in place

– How some IMs achieve such a low failure

frequency

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for power supply assets is 55

per thousand main track-kilometre and year

65

Power supply failures in relation to network size


• The failure frequency varies widely between

16 and 630 failures per thousand main track-

kilometre and year

• Power supply failure frequency plays a minor

role (7%) compared to the entire asset failure

rate in the network

• A more precise comparison would have to

take the degree of electrification into account,

i.e. the power supply failure rate should refer

to the length of electrified main track

• Especially in the power supply system, the

impact of asset failures on train operations is

essential (as already identified in the

dimension "Reliability")

KPI 54

55,3

0 50 100 150 200

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for track assets is 86 per


66

Track failures in relation to network size


• The failure frequency varies widely between 2

and 200 failures per thousand main track-

kilometre and year

• Five out of ten IMs (Adif, BaneNOR, Infrabel,

LG and SCNF R.) achieve a track failure rate

well below the weighted average

• PKP PLK, ProRail, SBB and TRV show a

decrease in 2017 compared to the average of

2012-2017

• As the track system is the second highest

failing asset group (12% of all asset failures),

an in-depth analysis could identify

– The main reasons for track failures

– How track quality is measured at the IMs

– If there is a correlation between track

failures and age of track (in terms of years

and/or accumulated gross tonnage)

KPI 55

86,0

0 100 200 300

Adif

Bane NOR (E)

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for structures is 1 per


67

Structure failures in relation to network size


• The failure frequency varies between 0 and

12 failures per thousand main track-kilometre

and year

• The higher number of failures at ProRail

results from the large number of bridges and

collisions by boats

• A few IMs show a slight decrease in 2017

compared to the average of 2012-2017

• Compared to the average asset failure

frequency in the peer groups’ network,

structure failure rates are negligible (0,1%)

• Similar to the power supply system, a more

precise comparison would have to take the

share of structures into account, i.e. the

structure failure rate should refer to the length

main track on bridges/ in tunnels

• Also for structures, the impact of asset

failures on train operations is essential

KPI 56

1,40

0 5 10 15 20

Adif

Bane NOR (E)

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB (D)

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

Average failure frequency for other assets is 89 per


68

Other infrastructure failures in relation to network size


• The failure frequency varies between 7 and

340 failures per thousand main track-

kilometre and year

• Some IMs appear to have a significant share

of other asset failures

• Compared to the average of 2012-2017, the

failure frequency for other assets was

increasing at PKP PLK

• For a meaningful interpretation, it needs to be

analysed further

– What is behind the other asset failures

– What are the reasons for the high failure

rates

KPI 57

89,2

0 100 200 300 400

Adif

Bane NOR

DB

FTIA

Infrabel (D)

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

The peer group’s weighted average for tracks with per-

manent speed restrictions is 2% of main track-kilometre

69

Tracks with permanent speed restrictions

% of main track-km (2017)

• Based on the definition, all permanent speed

restrictions that are already included in the

annual timetable should be provided

• Some IMs do not count permanent speed

restrictions at all, as these are included in the

working timetable

• It would be interesting to understand why

some IMs do not count PSR

• Additional value would be provided by a root-

cause analysis for PSR (e.g. postponed

renewals, lack of resources …)

• Furthermore, it would be interesting to

understand the difference between average

track design speed (in terms of track

standard) and average operated train speed

• It could be worth testing a new indicator

measuring this difference

KPI 58

1,79

0 2 4 6 8

Adif¹⁾Bane NOR

DB

FTIA¹⁾ (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

1) Data years: Adif 2016, FTIA 2015

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average, about 1,5% of the main track has temporary

speed restrictions due to deteriorating condition

70

Tracks with temporary speed restrictions


• While some IMs have hardly any TSRs,

others temporarily restrict speed on 6% of

their network

• An in-depth analysis could identify

– The statistical distribution of length and

duration of TSRs

– The reasons for temporary speed

restrictions (e.g. bad track geometry …)

• It would be also interesting to understand the

impact of TSRs on train operations

KPI 59

1,50

0 5 10

Adif¹⁾Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB¹⁾SNCF R.

TRV

1) Data of 2016

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Delivery – drivers1)

71



Delivery

Network

availabilityAsset condition+

Asset

management+

Planned unavail-ability (mainte-nanceetc.) (KPI43)

Speed restrictions (KPI 58/59)

Renewal rate

Maintenance regime

Age distribution

Activity rates

Activity unit costs

Asset downtimes

Capacity

Possessions planned


Condition

Asset failures

Signalling

Telecom

Power supply

Track

Structures

Other

Permanent speed

restrictions

Temporary speed

restrictions

Delivery

PRIME KPIs2)



Network

utilisation

Train frequency (KPI 92)

Traffic complexity

+

Innovation/ digitalisation

43

44

51

52

53

54

55

56

57

58

59

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Further analysis should focus on the concepts and the

impact of capacity constraints on train operations

Delivery – Further analysis

• Information on possession management is relatively sparse so far

• It would be worth understanding the different concepts of possession management in general

• Furthermore, a good practice exchange should focus on how IMs intend to optimise their

utilisation of possessions for maintenance, renewals and enhancements

• Concerning asset failure frequencies, it would be interesting to understand the reasons/ the

background for the wide range of frequencies among the peers, such as asset condition,

maintenance regimes, different failure recording technologies etc.

• The signalling system accounts for approximately two thirds of all asset failures

• Even if failure frequency is much lower, the impact of failing power supply, structure and

telecommunication assets on train delays is significant as already identified in the performance

chapter

• In order to identify the consequences of asset failures, their impact on train operations (i.e. train

operations as well as passengers or freight customers) would need to be analysed further

• Current available data on speed restrictions is improving but still sparse

• It would be beneficial to understand the different concepts, the drivers or main causes when to

set up either a temporary or a permanent speed restriction

• Similar to asset failures, also the impact of restricted network availability (by speed restrictions) on

train operations would need to be analysed further

72

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context


– Performance

– Delivery

– Financial

– Growth

• Appendix

73

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Support delivery of a cost-effective railway, through

identification and implementation of good practices and

processes;

• Identify and encourage opportunities to increase revenues

from all sources;

• Understand the impact of charging and charges on IM and

the whole railway industry; and

• Support making the case for appropriate and effective

investment in the railway.

Financial dimension is intended to provide understanding

of the structure and the level of costs and revenues

Financial – objectives

74


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

All financial data have been adjusted for purchasing power

and converted into Euro using purchasing power parities

PPPs1)

75

1) Data provided by European Commission

Country Currency 2012 2013 2014 2015 2016 2017

Belgium EUR 1,10 1,11 1,10 1,08 1,10 1,10

Finland EUR 1,21 1,24 1,24 1,22 1,24 1,24

France EUR 1,12 1,11 1,10 1,08 1,10 1,10

Germany EUR 1,04 1,05 1,04 1,03 1,06 1,07

Great Britain GBP 0,92 0,94 0,94 0,91 0,95 0,98

Italy EUR 1,00 1,01 1,01 0,98 0,99 0,99

Latvia EUR 0,67 0,68 0,68 0,67 0,67 0,69

Lithuania EUR 0,60 0,60 0,60 0,60 0,62 0,63

Netherlands EUR 1,10 1,09 1,09 1,09 1,10 1,12

Norway NOK 11,95 12,26 12,56 12,86 13,71 13,98

Poland PLN 2,40 2,41 2,41 2,36 2,40 2,47

Portugal EUR 0,78 0,79 0,78 0,78 0,80 0,81

Spain EUR 0,91 0,91 0,90 0,89 0,90 0,90

Sweden SEK 11,52 11,81 11,99 11,99 12,28 12,51

Switzerland CHF 1,79 1,79 1,75 1,67 1,69 1,68

Purchasing power parity (LCU/EUR)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The Costs category includes all the costs incurred by the IM, broken down into useful and comparable sub-categories. It includes all Operating, Capital and

Investment costs. For purposes of comparison, costs will be adjusted where appropriate to reflect local costs using purchasing power parities (PPPs). The

costs incurred by an IM will be dependent on a number of factors: some within and some outside the management responsibility of the IM.

Financial – Costs – Overview

76

PRIME


Total IMs annual operational expenditures per main track-km

Total IMs annual capital expenditures per main track-km

Total IMs annual renewal expenditures per main track-km

OPEX – operational

expenditures in relation to

network size

CAPEX – capital expenditures

in relation to network size

Renewal expenditures in

relation to network size

Total IMs annual maintenance expenditures per main track-kmMaintenance expenditures in

relation to network size

Total IMs annual traffic management expenditures per main track-kmTraffic management

expenditures in relation to

network size


Costs

OPEX

Maintenance

Traffic management

CAPEX

Renewals

Revenues

Non access charges


Proportion

Incentive regimes

60

62

66

68

64

87

81

91

80


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

1) Results are normalised for purchasing power parity

(accu-

racy)IM



Average annual operational expenditures are 88

thousand Euros per main track-kilometre

77

OPEX – operational expenditures in relation to network size1)

1.000 Euro per main track-km (2017)

• Operational expenditures vary between 41

and 217 thousand Euros per main track-

kilometre and year

• OPEX appear to be relatively constant over

time except for LG, showing a decrease in

2017 compared to the average of 2012-2017

• This comparison provides an overview about

annual expenditure levels independent of

different operational conditions, representing

major cost drivers

• For a meaningful gap analysis, these cost

drivers should be taken into account, e.g.

– Network characteristics (i.e. asset

densities)

– Network utilisation (i.e. train frequencies,

gross tonnage)

– Traffic management technologies and

degree of centralisation

KPI 60

87,7

0 50 100 150 200 250

Adif (E)

Bane NOR

DB

FTIA²⁾ (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

2) Data of 2015

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Maintenance and traffic management cover a significant

share in total operational expenditures

OPEX – operational expenditures in relation to network size1)


78


0 50 100 150 200 250

FTIA

ProRail

DB2)

Adif

Infrabel

TRV

Bane NOR

IP

LDZ3)

LG

SBB

NR2)

PKP PLK3)

RFI

SNCF R.

87,7

Traffic ManagementMaintenance Residual OPEX such as power consumption and other OPEX1) Results are normalised for purchasing power parity

2) Traffic Management not available, therefore included in residual OPEX

3) Disaggregation not available

• All 15 IMs provided total annual operational

expenditures

• 13 IMs provided annual maintenance

expenditures

• 11 IMs also provided annual expenditures for

traffic management

• Based on these 11 IMs, maintenance

accounts for 50% and traffic management

accounts for 20% of total operational

expenditures on average

• As the residual OPEX are about a third of

total OPEX, it would be worth analysing these

more in detail

• The weighted average of OPEX is 88


KPI 60

Total weighted average of total OPEX

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



Average annual maintenance expenditures are 38


79

Maintenance expenditures in relation to network size1)


• The range of maintenance expenditures

varies between 20 and 87 thousand Euros

per main track-kilometre and year

• With one major exception (LG), maintenance

expenditures appear to be relatively constant

over time

• Similar to the total expenditure level, the

comparative analysis of maintenance

expenditures should also take into account

major cost drivers such as network

characteristics and utilisation

• An in-depth analysis should further

differentiate

– Asset groups (track, signalling …)

– Preventive and corrective activities

KPI 62

38,1

0 50 100

Adif (E)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB (D)

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



Average annual expenditures for traffic management are

17 thousand Euros per main track-kilometre

80

Traffic management expenditures in relation to network size1)


• The range of expenditures for traffic

management varies between 4 and 27

thousand Euros per main track-kilometre and

year

• Similar to maintenance, also traffic

management expenditures appear to be

relatively constant over time except for

Infrabel and LG, showing a decrease in 2017

compared to the average of 2012-2017

• Operational expenditures for traffic

management are assumed to be driven

mainly by labour costs (as expenditures for

signalling assets are covered in maintenance

or CAPEX)

• An in-depth analysis should consider different

signalling technologies currently in use and

the varying degrees of centralisation

KPI 64

16,7

0 10 20 30 40

Adif (E)

Bane NOR (P)

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



On average 119 thousand Euros per main track-kilometre

and year are spent on capital expenditures

81

CAPEX – capital expenditures in relation to network size1)


• The range of annual capital expenditures

varies between 17 and 227 thousand Euros

per main track-kilometre and year

• In many cases, capital expenditures are

linked to major (re-) investment programs

• Thus it is not surprising that some IMs show

high fluctuations in expenditure levels over

time

• For an in-depth analysis, major cost drivers

should be taken into account such as

– Age and condition of the infrastructure

assets

– Technological migration strategies (such as

ERTMS)

– Available budgets and funding agreements

– Supplier market, prices and resources

KPI 66

119

0 100 200 300

Adif (E)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Renewal expenditures cover nearly 50% of the total capital

expenditures

CAPEX – capital expenditures in relation to network size1)


82


0 50 100 150 200 250 300

Adif

IP

DB

Bane NOR

NR

LG

Infrabel

FTIA

LDZ

PKP PLK

119

ProRail

RFI

SNCF R.

SBB

TRV

Renewal Enhancements, Investments & Other CAPEX


• 14 IMs provided both total annual capital

expenditures and renewal expenditures

• Based on these 14 IMs, renewal expenditures

accounts for 45% of total capital expenditures

• As the remaining expenditures, such as

enhancements and investments, are more

than half the total CAPEX, it would be worth

analysing these more in detail

KPI 66

Total weighted average of total CAPEX

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



Average annual renewal expenditures are 54 thousand

Euros per main track-kilometre

83

Renewal expenditures in relation to network size1)


• The range of renewal expenditures varies

between 5 and 126 thousand Euros per main


• Similar to the total CAPEX it is not surprising

that some IMs show high fluctuations in

renewal expenditure levels over time

• A constantly low renewal expenditure level

bears the risk of creating a reinvestment

backlog

• A high renewal expenditure level does not

necessarily mean inefficient renewal activities

• For a meaningful interpretation of results, the

varying stages within the entire life cycle of

the different asset groups need to be taken

into account

KPI 68

53,8

0 50 100 150

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB (D)

SNCF R.

TRV (D)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• The sum of annual maintenance and renewal

expenditures provide a snapshot of current

expenditures into the existing network

• As especially renewals are considerably

fluctuating over time, future analysis should

consider comparing multi-annual averages

• An individual gap analysis should further take

into account different operational conditions

and cost drivers that are outside or hardly in

control of IMs, such as

– Network complexity/ asset densities e.g.

switches, bridges, tunnels …

– Network utilisation e.g. train frequency,

gross tonnage

– Current stage of key assets/ asset groups

within the entire life cycle e.g. current

renewal rates compared to steady state

renewal rates

Maintenance and renewal expenditures in relation to network size1)


84


0 50 100 150 200

SBB

LG

Adif

LDZ2)

DB

Bane NOR

FTIA

IPInfrabel

NRPKP PLK

ProRail

RFI

SNCF R.

TRV

92,3

Renewal expenditures relative to network sizeMaintenance expenditures relative to network size

Average annual expenditures for maintenance and renewal

are 92 thousand Euros per main track-kilometre


2) Maintenance expenditures not available

KPI 62+68

Total weighted average of sum of maintenance and renewal expenditures

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

The Revenue category provides a summary of the total non-track access revenue ‘earned’ by an IM, excluding subsidies and property development.

Furthermore, it measures and compares that element of an IM’s revenue that comes from charges from operators using its network and service facilities.

Financial – Revenues – Overview

85

PRIME


Total IMs annual TAC revenues (including freight, passenger and touristic trains) compared to total main track-kmTAC revenue in relation to

network size

Total IMs annual income from incentive/performance regimes with customers (if applicable, no public grants or state

subsidies) per main track-km

Income from incentive

regimes in relation to network

size


Costs

OPEX

Maintenance

Traffic management

CAPEX

Renewals

Revenues

Incentive regimes

Non access charges


Proportion

Total IMs annual revenues from non-access charges (e.g. commercial letting, advertising, telecoms but excluding grants or

subsidies) related to total main track-km

Total revenues from non-

access charges in relation to

network size

Percentage of IMs annual TAC revenues (including freight, passenger and touristic trains) compared to total revenuesProportion of TAC in total

revenue

60

62

66

68

64

87

81

91

80


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



• The range of TAC revenues in relation to

network size varies between 6 and 170 thou-

sand Euros per main track-kilometre and year

• TAC revenues appear to be relatively

constant over time

• This KPI illustrates the degree to which IMs

manage to generate user revenues to cover

the cost of the network. The degree to which

IMs generate revenues from the utilisation of

the network by operators is provided by

relating TAC revenue to the traffic volume

(additional KPI 82)

• An in-depth analysis could focus on

– Track access charge regimes

– Differentiation into/ share of train types

• A more precise definition of TAC revenue and

its constituents will be provided in the future

Average annual revenues from track access charges are


86

TAC revenue in relation to network size1)


KPI 87

60,7

0 50 100 150 200

Adif (E)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB (D)

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



• TAC revenue in relation to traffic volume

appears to be more homogeneous among the

peer group than TAC revenue in relation to

network size (c.f. KPI 87)

• The range of TAC revenues in relation to

traffic volume varies between below 1 and

more than 20 Euros per train-kilometre and

year

• TAC revenues appear to be relatively

constant over time

Average annual revenues from track access charges are

4 Euros per train-kilometre

87

TAC revenue in relation to traffic volume1)

Euro per train-km (2017)

KPI 82

4,26

0 10 20 30

Adif (E)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB (D)

SNCF R.

TRV

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM • Five out of twelve IMs generate a proportion

of total revenue from track access charges

above the weighted average

• Three IMs realize about 50% of their

revenues by track access charges

• Adif, Bane NOR and RFI have increased their

proportion in 2017 compared to the average

of 2012-2017

• Total revenues excluding grants and

subsidies

Track access charges account for 79% of the total

revenues on average

88

Proportion of TAC in total revenue

% of monetary value (2017)

KPI 81

79,4

0 50 100

Adif (E)

Bane NOR

DB

FTIA¹⁾Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

1) Data of 2016

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



• Three out of 14 IMs manage to generate

above average revenues from non-access

charges (Adif, NR and SBB)

• Thus it would be interesting to understand in

detail, how IMs achieve these revenues and

what they are based on

• SBB’s above avg. revenues stem from

providing goods (e.g. switches, rails,

sleepers) and services (e.g. use of IT tools) to

other IMs and RUs in Switzerland

• Total IMs annual revenues from non-access

charges include commercial letting,

advertising, telecoms but exclude station

access charges, income from energy supply,

grants and subsidies

Average annual revenues from non-access charges are


89

Total revenues from non-access charges in relation to network size1)


KPI 80

25,3

0 50 100

Adif (E)

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV • Concerning the definition it should be

ensured that income from energy supply

is not included

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/


(accu-

racy)IM



• Compared to the total volume of annual

expenditures and revenues, incentive

regimes play a minor role

• Six out of 12 IMs neither receive a bonus nor

pay a malus

• RFI and TRV receive larger bonuses, Bane

NOR receive the most by far

• ProRail is the only IM who regularly pays a

significant malus

• Since national incentive schemes are not

comparable it would be worth understanding

the different regimes and what criteria they

are based on

The average annual income from incentives of

representative peers is 95 Euros per main track-kilometre

90

Income from incentive regimes in relation to network size1)


KPI 91

• This KPI will be separated from costs

and revenues as an individual category

within finance0,095

-1,2 -0,6 0,0 0,6 1,2

Adif

Bane NOR (E)

DB

FTIA²⁾ (D)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

2) Data of 2015

3) Not part of weighted average

Outlier: 5,03)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Financial – drivers1)

91



Financial

Infrastructure

(Life Cycle) Costs+

Infrastructure

Revenues

MaintenanceTAC

Station access charges

Non-access charges

PRIME KPIs1)PRIME KPIs2)

Enhancement

Other OPEX

Traffic management

Power consumption

Investment

Renewal

Other CAPEX

Costs

OPEX

Maintenance

Traffic management

CAPEX

Renewals

Revenues

Incentive regimes

Financial

Charging regime

Network utilisation

Incentive / perfor-mance regimes

Charging regime

Station utilisation (by trains)



Non access charges


Proportion

Property management

Advertising

Energy supply

60

62

66

68

64

87

81

91

80

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Further analysis should focus on effectivity and efficiency

of expenses and revenues in order to identify good practice

Financial – Further analysis

• Financial data is nearly complete as it appears to be easy to access/ to provide for the IMs

• It needs to be stated clearly that all comparisons only provide levels of annual expenditures and

revenues with a wide range of individual results

• The comparisons do not provide any information neither on effectivity (how much was done) nor

on efficiency (how much did it cost)

• In order to identify good practice and to enable individual gap analyses, major cost drivers outside

the (immediate) control of an IM need to be taken into consideration or even normalised such as

– Network characteristics (asset densities)

– Network utilisation (train frequencies, gross tonnage)

– Current stage of key assets/ asset groups within the entire life cycle e.g. current renewal rates

compared to steady state renewal rate

• Furthermore, different operational conditions need to be taken into account such as

– Signalling technologies/ degree of centralisation

– Asset age/ condition

– Available budgets/ funding agreements

– Track access charge regimes/ track access pricing systems

• A financial task force has been established to improve the analysis on financial data; it will continue

its efforts to provide further insights into infrastructure managers‘ funding, e.g. grants.

92

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


– Context


– Performance

– Delivery

– Financial

– Growth

• Appendix

93

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• Improve the use of the overall capacity of the railway

network;

• Encourage modal shift to rail from road and air;

• Promote multi-modal transport integration;

• Understand and use new technology, such as ERTMS,

effectively and efficiently to support the objectives of the IM

and the integrated railway.

Aim is to describe the current / future network use /

technology, and integration with other transport modes

Growth – objectives

94


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Utilisation is an essential measure of the performance of an IM. One of the most important objectives for an IM is to use its infrastructure as effectively as

possible. This measure also distinguishes between passenger and freight traffic. Utilisation has a major impact on the ability of an IM to cover its costs and

the utilisation of the infrastructure will also affect the future performance (other KPIs) of the infrastructure, e.g. overall condition.

Growth – Utilisation – Overview

95

PRIME


Average daily train-km on main track (passenger and freight revenue service only, no shunting, no work trains) related to

main track-km

Average daily passenger train-km on main track (revenue service only, no shunting, no work trains) related to main track-

km

Average daily freight train-km on main track (revenue service only, no shunting, no work trains) related to main track-km

Degree of utilisation – all

trains

Degree of utilisation –

passenger trains

Degree of utilisation – freight

trains


Utilisation

Train-km

Passenger trains

Freight trains


Deployment today

Deployment 2030

Intermodality

Passengers at

accessible stations117

92

93

94

98

101


Intermodal stations 116

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

• On average each of the peer group’s railway

tracks is frequented by 38 passenger and

freight trains per day

• The utilisation of the peer groups’ railway

networks varies widely

• On average railway tracks are frequented

between 17 to 81 times per day

• Only LDZ and LG are frequented by more

freight than passenger trains

• Of course these figures do not provide any

information about the distribution of utilisation

in the network and across different types of

lines

• The reasons for this situation are manifold

and should be further explored: the geogra-

phic characteristics of the country, its location

in Europe (transit countries), the quality and

acceptance of railway services etc.

The majority of the peer groups’ networks is frequented

by passenger trains

Degree of network utilisation – all trains

Daily train-km per main track-km (2017)

96


0 50 100

PKP PLK

DB

Infrabel

FTIA

ProRail

AdifBane NOR

IP

LG

38,3

LDZ

NR

RFI

SBBSNCF R.

TRV

Passenger trains Freight trains

KPI 92

Total weighted average of sum of all trains

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average each of the peer group’s railway tracks is

frequented by 32 passenger trains per day

97

Degree of network utilisation – passenger trains

Daily passenger train-km per main track-km (2017)

KPI 93

• The intensity of network use by passenger

trains ranges from 9 to 75 trains a day

• While most organisations show frequencies

between 22 and 39 trains, some like SBB and

ProRail use their networks more than

average

• Passenger traffic appears to be very constant

over time

• Recommended questions for further analysis:

– How is the utilisation distributed across

networks?

– To what extent are there congested and

significantly underutilised parts?

– Are there opportunities for a better use of

existing infrastructure?

31,5

0 50 100

Adif

Bane NOR

DB

FTIA¹⁾ (E)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

1) Data of 2016

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

On average 7 freight trains are running daily on each of

the peer group’s railway track-kilometres

98

Degree of network utilisation – freight trains

Daily freight train-km per main track-km (2017)

KPI 94

• While passenger trains on average use the

peer group’s network 32 times a day, the

figure for freight trains is more than four times

lower

• The different role of rail freight is expressed

by varying degrees of utilisation

• Freight traffic appears to be relatively

constant over time, except for decrease for

LDZ and an increase for ProRail

• What drives these results? Aspects for a

more in-depth analysis could be the service

offering in freight, its competitiveness against

other modes or the comparative size of the

infrastructure network

6,80

0 5 10 15

Adif

Bane NOR

DB

FTIA¹⁾ (E)

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

1) Data of 2016

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Asset capability describes the functionality of the IM’s railway network. It provides the overview of the capability of the network and specifically the extent

to which the network meets the TEN-T requirements. The asset capability describes the IM’s part of the interoperability of the peer group's railway network,

although it is recognised that achievement of interoperability requires capability and functionality from the railway operators as well.

Growth – Asset Capability & ERTMS – Overview

99

PRIME


Main tracks with ERTMS in operation in proportion to total main tracks (measured in track-km)

In 2030, the percentage of main track-km planned to have been deployed with ERTMS, i.e. main tracks equipped with both

- ETCS (European train control system; any baseline or level) and GSM-R (Global System for Mobile Communications);

and where ETCS and GSM-R are used in service

ERTMS deployment

Planned extent of ERTMS

deployment by 2030


Utilisation

Train-km

Passenger trains

Freight trains


Deployment today

Deployment 2030

Intermodality

Passengers at

accessible stations117

92

93

94

98

101


Intermodal stations 116

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

ERTMS is deployed on about 6% of all tracks of the peer

group's railway network

100

ERTMS track-side deployment


KPI 98

• The IM’s implementation strategies are

heterogeneous which is reflected in no

ERTMS deployment in some countries vs. a

high share in others of more than 20%

• In this sample Infrabel is showing the most

dynamic development, being one of the few

countries in Europe that has opted for a

nationwide roll-out of ERTMS

• The motivation to deploy ERTMS is different

(capacity, safety, obsolescence etc.) and

should be explored further to understand the

dynamics of implementation in the context of

the EU deployment plan

• Decreasing values over time are due to

added main track-kilometre without ERTMS

6,49

0 5 10 15 20 25

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR¹⁾PKP PLK

ProRail

RFI

SBB

SNCF R.

TRV

1) Data of 2015

98 (scale shortened

for readability)

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/



(accu-

racy)IM

By 2030 ETCS is expected to cover about 29% of the peer

group’s railway network

101

Planned extent of ERTMS deployment by 2030

% of current main track-km (2017)

KPI 101

• Also with respect to the future development,

the pattern remains heterogeneous

• Whilst some countries plan to equip the

complete network with ETCS, others show

more modest roll-out plans, ranging between

an extent of 0% to 110%

• On average ETCS is expected to be

implemented in about half of the peer group's

railway network by 2030

• BaneNOR and SBB: value greater than 100%

as the ETCS equipped network will be larger

than the current network; in a future version

of the report this could be improved by

introducing and using a new input “planned

main track-km in 2030”

29,4

0 50 100

Adif

Bane NOR

DB

FTIA

Infrabel

IP

LDZ

LG

NR

PKP PLK

ProRail

RFI

SBB

SNCF R. (D)

TRV

105%

1) Data of 2016

101%

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

A highly functional intermodality between different transport modes can bring traffic and business to the rail network. Since trains rarely offer a door-to-

door solution, and rather is a part of the mobility chain, connections between modes become essential for the customers. Intermodality promotes

efficiency for both freight and passenger traffic. Intermodality also increases the number of potential customers for rail.

Growth – Intermodality – Overview

102

PRIME


Percentage of public passenger railway stations with connections to public urban transport (metro, bus, tramways, light rail,

ferries etc.…) within the entire railway infrastructure network, independent of ownership (Source "Passenger stations":

European Commission, RMMS)

Percentage of passengers registered annually in all accessible stations within the entire railway infrastructure network,

independent of ownership, related to the total number of passengers. An accessible station is one on which a passenger

can, from entering the station, reach the platform via level-access, without steps or equivalent.

Intermodal stations

Passengers using accessible

stations


Utilisation

Train-km

Passenger trains

Freight trains


Deployment today

Deployment 2030

Intermodality

Intermodal stations

Passengers at

accessible stations

116

117

92

93

94

98

101


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Growth – drivers1)

103



Growth

Offered capacity (train-km)

Track design speed

Competitive price (TAC) compared to other modes

RA(M)S (H)E

Intermodal stations/terminals

Traffic management

PRIME KPIs2)

Utilisation

Train-km

Passenger trains

Freight trains


Deployment today

Deployment 2030

Intermodality

Intermodal stations

Passengers at

accessible stations

Growth



Congestion on roads

Network utilisation Modal share of rail+Use of innovative

technologies+

Innovation strategy (e.g. digitalisation)

Age of current technologies (e.g. signalling)

Status/inspiration from other industries

Freight/passenger traffic mix 116

117

92

93

94

98

101

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Further analysis should account for underlying root causes

and identify opportunities for improvement

Growth – Further analysis

• From a total network perspective the utilisation of European railway infrastructure varies

significantly; in order to better understand to what extent parts of the networks are over- or

underutilised a drill-down into the distribution of utilisation would be valuable

• Furthermore, discussion with IMs should be started to investigate the activities undertaken to

manage capacity, for example by increasing the use of existing infrastructure or downsizing parts

of the network

• The development of freight traffic over the years is quite different, too. Some countries face a slight

increase while others remain stable or even run less trains per day. To understand these

developments the drivers should be analysed as well as the activities that IMs have undertaken to

increase the attractiveness of rail freight

• From country to country the motivation to roll out ETCS can be different (capacity, safety,

obsolescence etc.) and should be explored further in order to understand the different levels and

dynamics of its implementation

• Signalling failures account for the majority of infrastructure related unreliability and in general

ERTMS provides the opportunity to reduce them; however, the correlation is not quite clear and

further analysis would be helpful to understand the trade-offs between the signalling system

and reliability

• With regards to KPIs on intermodal stations and passengers using accessible stations more effort

is needed to complete these datasets

104

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Table of contents

• Introduction


• Appendix

105

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Country characteristics & market and operations

Contextual information – Countries (2017)

106

Source: civity calculations using data as provided by the infrastructure managers until 29 January 2019 unless otherwise noted

1) EC RMMS; 2) IRG Rail; 3) TENtec database; all provided by the European Commission, 03 January 2019; 4) RFI (25/02/19)

Sp

ain

No

rway

Germ

an

y

Fin

lan

d

Belg

ium

Po

rtu

gal

Latv

ia

Lit

hu

an

ia

Un

ited

Kin

gd

om

Po

lan

d

Neth

erl

an

ds

Italy

Sw

itzerl

an

d

Fra

nce

Sw

ed

en

Country characteristics

Country area (thousand km2) 506 324 357 338 31 92 65 65 244 313 42 301 4 633 450

Population (million) 47 5 83 6 11 10 2 3 66 38 17 61 8 67 10

Currency EUR NOK EUR EUR EUR EUR EUR EUR GBP PLN EUR EUR CHF EUR SEK

GDP per head (index - EU28 100) 93 146 124 109 117 77 67 78 105 70 128 96 161 104 122

Number of border countries 6 3 9 3 4 1 4 4 1 7 2 6 5 8 2

Population density (persons/km2) 92 16 231 16 372 112 30 44 270 121 411 201 204 106 22

Market and operations (national)

Number of RUs1) (2016) 38 7 448 3 7 10 6 12 41 82 39 33⁴⁾ 47 21 32

Share of NW managed by main IM2) 100% 100% 85% 100% 100% 100% 100% 100% 97% 96% 100% 84% 59% 100% 89%

% of main lines in TEN-T core network3) 54% 0% 25% 22% 34% 57% 53% 45% 22% 25% 28% 33% 0% 31% 36%

Modal share of rail freight 5% 13% 19% 27% 11% 15% 77% 65% 8% 25% 6% 15% 38% 11% 29%

Modal share of rail passengers 7% 5% 9% 6% 8% 4% 4% 1% 9% 7% 11% 6% 17% 10% 9%

% of freight in total train-km 13% 15% 30% 19% 16% 49% 58% 7% 32% 7% 13% 16% 14% 23%

% of international in passenger-km1) (2016) 1% 1% 5% 3% 4% 3% 7% 33% 2% 3% 7% 1% N/A 12% 4%

% of international in tonne-km1) (2016) 18% 46% 48% 35% 78% 8% 98% 74% 0% 43% 91% 50% N/A 30% 36%

Countries

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Organisation & Network

Contextual information – Infrastructure Managers (2017)

107

AdifBane

NORDB FTIA

Infra

belIP LDZ LG NR

PKP

PLK

Pro

RailRFI SBB

SNCF

R.TRV

Organisation

Is the IM state-owned Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Are IM and operators integrated No No No No Yes No No No No No Yes Yes No

Number of FTE employees 12.079 4.420 639 3.697 6.494 2.987 38.594 39.349 4.280 25.963 53.624 7.135

Age average 53,7 45,0 48,5 49,6 46,0 46,0 46,0 46,7 48,8 41,2 47,0

Male employees among IM's workforce 84% 76% 60% 76% 64% 68% 68% 74% 88% 88% 62%

Network

Main line km (lines in commercial use) 15.302 3.856 5.926 2.546 1.860 18.513 3.169 16.787 28.120 9.676

Total main track-km 21.029 4.125 55.311 6.708 6.515 3.244 2.217 1.911 31.221 27.120 5.409 27.044 6.183 48.992 11.775

Total passenger high speed main track-km 5.248 0 1.527 0 0 0 0 294 2.428 4.401 0

Single track-km per total track-km 37% 85% 61% 53% 47% 77% 28% 15% 31% 20% 54%

Electrified track-km per total track-km 60% 60% 69% 59% 86% 75% 16% 9% 69% 77% 77% 100% 56% 84%

Million train-km 197 48 1.022 48 105 36 14 15 572 235 160 354 178 475 157


12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

We used a scoring model to evaluate the overall

robustness of the KPIs across all IMs

Methodology: Robustness traffic light

108

Measure Calculations Robustness outcome

For each measure, IMs

evaluate robustness

levels. Weightings were

assigned to each level.

Depending on the number

of evaluations for each

level a total score was

calculated.

The total score was

compared to the maximum

score to determine the

robustness outcome.

1 2 3

Level Weighing

Count for each

level (examples)

Score for

each level

“Normal”

“Estimate”

“Deviating from

definition”

“Preliminary”

4

2

1

1

5

2

0

1

20

4

0

1

=

=

=

=

Total

score

∑ 25

“Maximum score1)

depends on the number of

evaluations”

•

•

•

•

1) The maximum score implies that the level of all provided robustness evaluations was “normal”

75%

100%

50%

32

24

16

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

Editors

109

Klaus Wittmeier

Große Reichenstraße 27

20457 Hamburg

phone: +49 (0)40 181 22 36 67

mobile: +49 (0)160 706 32 25

[email protected]

www.civity.de

Willi Flemming


20457 Hamburg

phone: +49 (0)40 181 22 36 50

[email protected]

www.civity.de

Frank Zschoche


20457 Hamburg

phone: +49 (0)40 181 22 36 66

mobile: +49 (0)171 771 17 90

[email protected]

www.civity.de

12

10

01

06

_P

RIM

E_

Be

nch

ma

rkin

g_

Re

po

rt_

Pu

blic_

Ve

rsio

n_

se

nt_

to_

su

bg

rou

p_

20

19

05

03

.pp

tx©

civ

ity 2

01

9//

/

110

good practice benchmarking of the rail infrastructure managers · network rail nr gb benchmarking...

Documents